Making the Case for Data Journalism in Our Community College Classrooms

Graph on data driven journalism including data, filter, visualize, storyAnyone who has read my blog at all knows that data journalism is something I’ve been thinking quite a bit about for the past 3 or 4 years. I am no expert; in fact, I often feel like a complete phony documenting my “growth” in this area, since there is so much more for me to know. But that’s all the self-flagellating I’m going to do. You know why? Because at least I’m trying.

I made the case to members of the Journalism Association of Community Colleges a few days ago at our summer faculty conference. True to form, the participants were supportive and agreeable. I expected that. After all, it is through this organization that I got a jump start learning about audio, video, and social media. The organization is full of passionate journalism teachers, teachers whose battle cry is definitely NOT “but it’s always been done this way!” These are professionals with our students’ best interest at heart, and that means they are willing to learn and tackle whatever is coming down the pike in our field.

So it was a pleasure to present. Again, I’m no expert, but I did my best to synthesize the information I’ve learned from taking a number of MOOCs on data journalism (the most recent of which was Data Journalism Fundamentals) and experimenting with data units in my classes. For those knowledgeable about data journalism, these will not rock your world, but like I told my audience today, it’s a start. Here’s what I talked about.

Why we need to do more to bring data journalism into our community college journalism courses: I cited a Knight Foundation-funded study from that came out recently about the state of data journalism in American universities. This study found that out of the 113 AEJMC accredited journalism programs, fully 54 had not a single data journalism course. In this era of big data and big data stories (think Panama Papers, for starters), that’s a problem. And we are in an era not just of big data, but also open data. That should mean that journalists are now irrelevant, right? job boards at nicarActually, even though audiences can now access data they could never have directly seen in the past, they lack the time and expertise to make this huge amount of information meaningful. That’s, points out Simon Rogers, where journalists come in. And, in fact, data journalism is the one of the few bright spots in the journalism job market.

The skills we need to teach our students: Everything I’ve learned so far suggests to me that there are five essential skills I want my students to leave with.

  • Skill #1: Data Acquisition–My students need to know how to a) use advanced search techniques, b) scrape a website (even if it’s mediated through easy-to-use tools), and c) file a FOIA request. I also want them aware of some of the big data repositories out there.
  • Skill #2: Backgrounding Data–We would never send our students out to interview the basketball coach at our college without minimal research, such as whether or not the team is having a winning season. It’s the same with data. Our students need to learn to ask questions not just of the data, but about the data. Who collected the data? How did they collect it? How often? Why did they collect it? Our students need the confidence to move beyond trusting the data just because the government or an expert collected it.
  • Skill #3: Cleaning Data–Data is collected by people, and people are imperfect Ergo, expect imperfect data. Our students need to know how to use sorting and filtering to clean data. If you are more ambitious about this whole thing, go ahead and teach them some OpenRefine. But our students should feel comfortable and even empowered by having some spreadsheet knowledge.
  • Skill #4: Interviewing Data–Again, the power of spreadsheets shall reveal all. Or at least, reveal things we could never see just by eye-balling the whole document. Let’s show our students how they can chain filter and slowly move in on the data and see things we could not see before. In our most introductory courses, let’s make sure students understand how to use basic spreadsheet calculations. Once they get that, let’s show them how to use pivot tables.
  • Skill #5: Visualizing Data–We need to introduce our students to data visualization concepts. The goal is to help them think about data visually and to make data visualizations that are accurate, meaningful, useful and maybe even beautiful. Sure, it may start with an Excel graph, but can easily move onto to a Google Fusion Table or a visualization made in Silk. We may not have the resources at our schools to create New York Times-style interactive extravaganzas, but that should not stop us from creating what we can.

Possibilities for our classes: In my workshop, I shared the three-week data journalism unit I use in my JOUR 121: Advanced Writing and Reporting class. Not ready for that? No problem. Just don’t let “overwhelm” stop you from doing something, anything. For example, how about using your basic news writing course to throw in a few advanced search skills? Or teach a few basic Excel calculations? In your production classes, make sure students now how to make simple, interactive Google maps. Start there. When you are comfortable, introduce other things.  Still not sure how to incorporate data into your program? The Knight Foundation-funded study I linked to above, Teaching Data and Computational Journalism, makes some great suggestions for our level, starting on page 53.

Resources: The good news is that there is so much help out there, from MOOCs to websites to inexpensive or free handbooks. Again, baby steps. I’ve posted some of my favorites on the website I created for the workshop.

Let me know how you plan to add data journalism into your program. Our California Community College journalism programs have worked hard to update curriculum and stay current. Let’s continue that now with data journalism skills.



Incorporating Data Journalism: Finally a Victory

I’ve written about my mission to get data journalism and visualization into my JOUR 121 course before. Not being a math person myself (yeah, another journalist who has math phobia–blah, blah, blah), I’ve taken a variety of MOOCs and read a lot to try to get myself up to speed.

I’ve struggled for semesters to figure out the right assignment for my students. At first, I was appropriately underambitious, creating a short Excel spreadsheet of fake data for them to play with and write a fake story with. It wasn’t particularly intriguing or edifying. I was just out of the gates.

Next–and by next, I mean probably 5 semesters–came overly ambitious assignments that I composed under the heady inspiration of all that I was learning. The problem? I simply could not help them when they ran into the inevitable problems. That’s what I got for trying to have them scrape, background, clean and interrogate huge data sets and ready it all for a Google Fusion Table intensity map. When things didn’t work, I’d have to water down the assignment.

All good intentions, I assure you, and I am proud of myself for getting ahead of myself on this one–because it was necessary to start.

But I can finally cop to some success. This semester, I ditched the traditional story that I attached to each data assignment. Instead, I had them once again look at Clery Act data, but this time I limited the scope to disciplinary action violations for one year. And I took a page out of Steve Doig’s module in the MOOC, Data Driven Journalism, by providing my own video tutorials and walking them step by step through backgrounding to interrogation, using an older data set. The students interviewed our chief of public safety to help them background the data. I then “quizzed” the students in a way that gave them the opportunity to go through the same process with data from a different year. Finally, the students made Google Fusion Table maps with the schools that had the highest number of disciplinary action violations.  The students also came up with a variety of story ideas based on what the data revealed. They were successful with all of this.

I will take the victory.

Would I ultimately like them to write the stories connected to the data? Yes, of course. But with these beginners, this is enough. For now. And because I have also finally started bringing this necessary training into our other classes, particularly the publication class, the students are beginning to think about real data stories and how to visualize them.

Google Fusion Tables Fail (But I’ll Keep Trying)

I’ve been trying to learn how to have my students mash up two data sets and map them through Google Fusion Tables. Every tutorial I click on to learn more about Google Fusion Tables tells me how easy this is.

Yeah, no.

OK, let me modify that a bit.  I can get a data set into Google Fusion easily.  On occasions, I’ve even gotten Fusion Tables to locate my data on a map.  But I’ve also run into myriad problems.  And sure, this is to be expected, given that I am such a novice in the world of data-driven journalism.  But Lord knows I’m trying. I subscribe to a number of data-driven journalism feeds, recently took the Knight Center Data-Driven Journalism: The Basics and am signed up for Doing Journalism with Data: First Steps, Skills and Tools.  I’ve viewed dozens of tutorials on YouTube and have experimented multiple times on Fusion.

So, what are the blocks I’ve encountered?

  • Getting two data sets to align in the “merge” option.
  • Once the files have merged, getting the data to show in a way that doesn’t replicate certain columns. (See photo.)
  • And what’s up with the map? (See photo #2.)
Exhibit #1: Problems with merged files

Exhibit #1: Problems with merged files

There’s so many small things that I don’t yet understand and I suspect it’s these small details that are causing me big problems. What I really need to do is to find someone with more expertise than I have with whom to work.

Exhibit #2: Failed attempt to properly map the data

Exhibit #2: Failed attempt to properly map the data

Any suggestions?